Benchmarking Retrieval-Augmented Large Language Models in Biomedical NLP: Application, Robustness, and Self-Awareness

2024-05-13

arXiv.org

AI Summary

Retrieval-augmented large language models (RALs) lack systematic evaluation in biomedical NLP, hindering assessment of their reliability and trustworthiness. Method: We introduce the first comprehensive RAL benchmark for biomedicine, comprising four core capabilities—unlabeled-knowledge robustness, counterfactual robustness, diverse-input robustness, and negative-knowledge awareness—supported by four dedicated testbeds. We conduct cross-model, cross-task, multi-dimensional evaluation across nine biomedical datasets and five NLP tasks, integrating three mainstream LLMs and three retrieval architectures, enhanced by an adaptive retrieval-augmentation mechanism and novel meta-evaluation metrics. Contribution/Results: Experiments reveal pervasive failures of RALs under counterfactual and negative-knowledge scenarios, exposing critical trust deficiencies. To bridge this gap, we publicly release the benchmark datasets, evaluation toolkit, and fully reproducible results—establishing a foundational resource to advance trustworthy biomedical RAL research.

Technology Category

• Natural Language Processing: Safety and Robustness
• Machine Learning: Calibration & Uncertainty Quantification
• Knowledge Representation and Reasoning: Reasoning with Beliefs

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Large language models (LLM) have demonstrated remarkable capabilities in various biomedical natural language processing (NLP) tasks, leveraging the demonstration within the input context to adapt to new tasks. However, LLM is sensitive to the selection of demonstrations. To address the hallucination issue inherent in LLM, retrieval-augmented LLM (RAL) offers a solution by retrieving pertinent information from an established database. Nonetheless, existing research work lacks rigorous evaluation of the impact of retrieval-augmented large language models on different biomedical NLP tasks. This deficiency makes it challenging to ascertain the capabilities of RAL within the biomedical domain. Moreover, the outputs from RAL are affected by retrieving the unlabeled, counterfactual, or diverse knowledge that is not well studied in the biomedical domain. However, such knowledge is common in the real world. Finally, exploring the self-awareness ability is also crucial for the RAL system. So, in this paper, we systematically investigate the impact of RALs on 5 different biomedical tasks (triple extraction, link prediction, classification, question answering, and natural language inference). We analyze the performance of RALs in four fundamental abilities, including unlabeled robustness, counterfactual robustness, diverse robustness, and negative awareness. To this end, we proposed an evaluation framework to assess the RALs'performance on different biomedical NLP tasks and establish four different testbeds based on the aforementioned fundamental abilities. Then, we evaluate 3 representative LLMs with 3 different retrievers on 5 tasks over 9 datasets.

Problem

• Evaluating retrieval-augmented LLMs' performance across biomedical NLP tasks
• Assessing robustness against unlabeled, counterfactual, and diverse knowledge
• Investigating self-awareness capabilities in biomedical retrieval-augmented systems

Innovation

• Retrieval-augmented LLMs retrieve pertinent biomedical database information
• Systematic evaluation framework assesses robustness across biomedical tasks
• Tests models on unlabeled, counterfactual and diverse knowledge scenarios